Build a Brain Tumor Image Segmentation Application
This reference implementation applies the U-Net architecture, a convolutional neural network, to segment brain tumors identified in raw MRI images.
|Target Operating System||Ubuntu* 16.04 LTS|
|Time to Complete||30 minutes|
|Software Used||Intel® Distribution of OpenVINO™ Toolkit|
Use the Intel Distribution of OpenVINO toolkit to detect brain tumors in MRI images. A pretrained model from open source datasets helps accurately predict results using the Sørensen–Dice coefficient.
Gain insight into the following solutions:
- Computer vision applications for healthcare
- Computer vision inference for medical image processing
Learn to build and run an application with these capabilities:
Perform segmentation of MRI scans to detect brain tumors.
Calculate the Sørensen–Dice coefficient and plot the prediction results.
Save output images that show segmented areas in MRI scans.
Using a combination of different computer vision techniques, this application performs brain tumor image segmentation on MRI scans and plots the Sørensen–Dice coefficient.
- Train the model using an open source dataset from the Medical Segmentation Decathlon for segmenting nerves in ultrasound images and lungs in computed tomography (CT) scans. More Information
- Use the model with the inference engine. Apply the results to calculate the Sørensen–Dice coefficient and to plot predictions from a segmented brain tumor.
- Store the output images from the segmented brain tumor locally.